In 1998, the Australian Research Council (ARC) funded a three year SPIRT project, with the then Queensland Department of Natural Resource and Mines (QDERM), to evaluate the potential of airborne Synthetic Aperture Radar (SAR) data for quantifying forest structure and AGB, thereby informing on the potential of JAXA’s ALOS-1 PALSAR (launched in 2006) for regional mapping of carbon stocks and forest growth stages and types. The 40 x 60 km area near Injune, the ILCP, was selected for the study as extensive clearance of vegetation occurred in the mid 1990s and structural formations were typical of many vegetation communities in Queensland (Tickle et al., 2006). AIRSAR multi-frequency (C, L and P-band) polarimetric data were acquired in 2000 as part of NASA’s Jet Propulsion Laboratory (JPL) PACRIM II mission (Lucas et al., 2004). To support interpretation of these data, hyperspectral Compact Airborne Spectrographic Imager (CASI), discrete return lidar (Optech ALTM1020) and 1:4000 scale colour aerial photographs were acquired over the same period, across a sample grid (10 columns x 15 rows) of 150 500 x 150 m Primary Sampling Units (PSUs; Fig. 1b), with each containing 30 50 x 50 m Secondary Sampling Units (SSUs). Forest inventories were conducted and allometric equations relating AGB to tree size measurements were generated for tree species common to the site.
The ILCP is recognised internationally as a key site (and one of only a few located in woodlands and open forests) that has facilitated the development of new algorithms for retrieving biophysical attributes and detecting change (e.g., in biomass, structure). The long term (2000 to 2012) remote sensing observations at multiple scales (ranging from individual trees to the landscape) have involved airborne and spaceborne sensors operating in different modes, with their interpretation supported by on-ground measurements at the tree and stand level. These data continue to provide an unprecedented opportunity to better understand ecosystem response to change but also unique opportunities for calibration and validation of new spaceborne sensors (e.g., the ALOS-2 PALSAR) and derived products (e.g., biomass maps). The lidar/optical acquisition of 2000 was also one of the earliest sets acquired over woodlands in Australia and provides an important baseline against which to detect change at a landscape scale. The 2009 lidar/optical data were acquired following an extreme drought, significant human-induced (clearing) and other natural change (fires). In each case, optical airborne datasets were also acquired, allowing interpretation of change. The time-series datasets collectively provide a unique record of change across scales (tree to landscape) that can be used to better understand and quantify ecosystem response to natural and human drivers. There is also an inevitable lag response, making acquisitions several years after these events particularly informative when interpreting change and hence future acquisitions are planned.
Research conducted to date has established that retrieval of above ground biomass from the L-band backscatter data is optimal when surface conditions are relatively dry and that the level of saturation with respect to biomass can be increased using such data.
In the north of Australia, both a seaward and landward movement of mangroves has been observed from time-series of Landsat sensor data and aerial photography.
Across much of Australia, inundation is seasonally variable and whilst areas of open water can be detected using optical remote sensing data (e.g., Landsat or MODIS), inundation beneath vegetation is often difficult to discern.